Librispeech(100h)
Emerging9papers using it
2022first seen
Librispeech(100h) is a dataset containing 100 hours of English speech used to evaluate the performance of automatic speech recognition (ASR) systems.
Papers using Librispeech(100h) (9)
- Effective and Efficient One-pass Compression of Speech Foundation Models Using Sparsity-aware Self-pinching GatesXNOR-FORMER: Learning Accurate Approximations in Long Speech
TransformersContrastive Siamese Network for Semi-supervised Speech RecognitionInvestigation of Ensemble features of Self-Supervised Pretrained Models
for Automatic Speech RecognitionDecoupled Structure for Improved Adaptability of End-to-End ModelsFastInject: Injecting Unpaired Text Data into CTC-based ASR trainingOne-pass Multiple Conformer and Foundation Speech Systems Compression
and Quantization Using An All-in-one Neural ModelSample adaptive data augmentation with progressive schedulingUniEnc-CASSNAT: An Encoder-only Non-autoregressive ASR for Speech SSL
Models